Foxxy Bridge
MCP Foxxy Bridge is a one-to-many proxy for the Model Context Protocol (MCP). It lets you aggregate and route requests to multiple MCP servers through a single endpoint, so you can centralize configuration for all your MCP servers.
README
MCP Foxxy Bridge
<!-- BADGIE TIME -->
<!-- END BADGIE TIME -->
<p align="center"> <img src="media/mcp-foxxy-bridge_logo_trimmed.webp" alt="MCP Foxxy Bridge Logo" width="300"> </p>
Overview
MCP Foxxy Bridge is a one-to-many proxy for the Model Context Protocol (MCP). It lets you aggregate and route requests to multiple MCP servers through a single endpoint, so you can:
- Centralize configuration for all your MCP servers
- Expose all tools, resources, and prompts from connected servers
- Route requests transparently to the right backend
- Use a single MCP endpoint in your AI tools (Claude Desktop, VS Code, etc.)
🚀 Quickstart
See the Installation Guide for full details.
1. Choose one of the following installation methods
A. Install via uv (Recommended):
uv tool install mcp-foxxy-bridge
B. Install latest from GitHub:
uv tool install git+https://github.com/billyjbryant/mcp-foxxy-bridge
C. Run with Docker (from GHCR):
docker run --rm -p 8080:8080 ghcr.io/billyjbryant/mcp-foxxy-bridge:latest --bridge-config /app/config.json
2. Run the Bridge
With config file:
mcp-foxxy-bridge --bridge-config config.json
Or with named servers:
mcp-foxxy-bridge --port 8080 \
--named-server fetch 'uvx mcp-server-fetch' \
--named-server github 'npx -y @modelcontextprotocol/server-github' \
--named-server filesystem 'npx -y @modelcontextprotocol/server-filesystem'
See Configuration Guide for config file examples.
3. Connect Your AI Tool
Point your MCP-compatible client to:
http://localhost:8080/sse
See API Reference for integration details.
📚 Documentation
- Overview & Features
- Installation Guide
- Configuration Guide
- Deployment Guide
- API Reference
- Architecture Overview
- Troubleshooting Guide
- Example Configurations
🛠️ Development
🤝 Contributing & Support
🔒 Security
MCP Foxxy Bridge follows security best practices:
Network Security
- Default binding: Bridge binds to
127.0.0.1:8080(localhost-only) by default - MCP server isolation: Individual MCP servers communicate via local stdio pipes, never network ports
- Configurable access: Host and port settings can be configured via config file or CLI arguments
Configuration Priority
- Command-line arguments (
--host,--port) - highest priority - Configuration file bridge settings (
bridge.host,bridge.port) - Secure defaults (
127.0.0.1:8080) - lowest priority
Security Recommendations
- Keep the default
127.0.0.1binding unless external access is required - If external access is needed, use proper firewall rules and authentication
- Regularly update MCP server dependencies
- Monitor server logs for unusual activity
⚖️ License
This project is licensed under the GNU Affero General Public License v3.0 or later (AGPLv3+). See the LICENSE file for details.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。